Introduction to TensorFlow on SeaWulf

TensorFlow is an open-source library for numerical computation and machine learning. It is widely used for developing and training deep learning models in fields such as computer vision, natural language processing, and reinforcement learning.

SeaWulf provides various versions of TensorFlow, both CPU and GPU optimized, allowing users to choose the best version based on their system requirements and workload demands. Below are the available TensorFlow modules across different versions.

Available TensorFlow Versions

SeaWulf provides the following versions of TensorFlow, categorized by GPU support:

CPU Versions

• tensorflow-cpu/latest (latest version for CPU)
• tensorflow/1.10
• tensorflow/2.4.1
• tensorflow/2.10

GPU Versions

• tensorflow-gpu/latest (latest version for GPU)
• tensorflow2-gpu/2.1.0
• tensorflow2-gpu/2.2.0
• tensorflow2-gpu/2.7.0

ROCm Version

• tensorflow-rocm/2.17.0 (for AMD GPUs using ROCm)

To load a specific version of TensorFlow, use the module load command: